Potential Safety Improvements in Rebreather Design

[Omisson]

So, after a student does a course with you, they are no longer capable of making an assembly error (which in a rebreather will lead to loss of consciousness and drowning)?

Would you feel more comfortable, despite your teaching skills and your student learning abilities, if, ultimately, the student at a later date would not be able to make an assembly error because the diving equipment (i.e. rebreather) rather than being designed as a Lego, it is designed such that:

“It shall not be possible to assemble or combine the components or parts in such a way that it can affect the safe operation and safe use of the apparatus, e.g. by incorrect connection of the hoses to the breathing circuit.”

???

Every diver makes mistakes (that should be the safest assumption... invincibility is for teens).

Jim didn't say people don't make mistakes. He said he does additional familiarization training for his students.

You just see what you want.

Rebreathers can never be 100% safe- nor can open circuit.

In your world open circuit regulators should be capable of detecting oxygen content and shut off when the MOD us reached.

Just silly.

 

[gianaameri]

Jim didn't say people don't make mistakes. He said he does additional familiarization training for his students.

You just see what you want.

Rebreathers can never be 100% safe- nor can open circuit.

In your world open circuit regulators should be capable of detecting oxygen content and shut off when the MOD us reached.

Just silly.

The rebreather manufacturers got together and wrote EN14143 to their liking to include:

“It shall not be possible to assemble or combine the components or parts in such a way that it can affect the safe operation and safe use of the apparatus, e.g. by incorrect connection of the hoses to the breathing circuit.”

That is what it is and not what I want or want to see and it was the outcome of a long process which the manufacturers agreed to collectively.

Silly, huh?

 

[TSandM]

Bear with me -- the first part of this post will seem irrelevant.

Many years ago, we lost a foal who was born prematurely. He went to the vet hospital, where they put him on oxygen, by putting a red rubber tube down his nose, and attaching some IV tubing to it. He also had an IV. After some hours, when he got strong enough to try to get up, they would disconnect both sets of tubing, let him try (they flail a lot in the process, which is why they disconnected everything), and then hook everything back up when he gave up and laid down. At 4am, they hooked the oxygen to the intravenous catheter and put 15 liter of gas into his bloodstream, and killed him.

One of our requests to the vet hospital was that they change their procedures so that this simply couldn't be possible (different tubing). Our Children's Hospital in Seattle did the same thing, when a neonates's tube feedings got hooked to an IV catheter and killed the baby.

Human error is not avoidable. It seems reasonable that, to the extent possible, things which are mission-critical ought to be designed so that improper assembly is not possible. It's not that hard to do -- just using incompatible fittings will prevent a lot of misassembly. Now, if this was a scrubber packing issue, that may be unavoidable. And if something was put together without an o-ring, or with a pinched o-ring, that may not be preventable, which is where pre-dive checklists come in.

 

[Omisson]

The rebreather manufacturers got together and wrote EN14143 to their liking to include:

“It shall not be possible to assemble or combine the components or parts in such a way that it can affect the safe operation and safe use of the apparatus, e.g. by incorrect connection of the hoses to the breathing circuit.”

That is what it is and not what I want or want to see and it was the outcome of a long process which the manufacturers agreed to collectively.

Silly, huh?

It is both engineering-wise and practically impossible to eliminate human error in the operation of ANY device. ANY device.

There is no completely self contained rebreather on the market that does this- no matter what is said. First sorb material must be exchanged each dive. Cells must be replaced bi-annually. Batteries and bulbs changed or charged routinely. BOVs must be cleaned, as must loop hoses and counter lungs. Computers must have firmware updates or software upgrades.

Unless there were multiple redundant monitors for all these things AND water seepage, AND multiple onboard co, co2 and O2 detectors, AND any anomalies found by those monitors caused a system shut down pre dive so as to render the unit inoperable - no one could say that standard in reality could be met.

You are just being.... You.

 

[Dr. Lecter]

Is that still the case if you reverse the left and right counterlungs and/or the one-way valves?

Any other ways it could be done in error?

Like I said, the ingenuity of fools knows no bounds. Incompatible inhale and exhale connections is a fine idea, so long as WOB and other function and reliability considerations are not harmed. Personally, I find color coding sufficient. When you start worrying about reversing whole CLs and/or BOV or DSV assemblies...you're taking it a bit far.

 

[rjack321]

3 minutes into the dive and their was something fundamentally wrong with that scrubber or assembly (like missing O-rings). Breakthrough or bypass that rapid should have been detected on pre-breath even.

---------- Post added November 27th, 2014 at 01:17 PM ----------

This case raises the point I raised in response. Why is the design such that it is apparently so easy for highly trained and experienced users to screw it up? I am a pretty careful diver, but I often jokingly refer to myself as the poster child of adult ADD. I figure if anyone is going to screw up a rebreather setup, it is going to be me. One of my friends who was new to rebreathers spoke of his pre-dive checklist once when we were diving together. IIRC, it had 43 steps. I thought, "Wow! Lots of opportunities for me to make a fatal error."

I hope, therefore, that those of you who are more knowledgeable about this than I, which is probably most of you reading this, will address this issue. How complicated is it really? Can it be made simpler? Will making it simpler make it less prone to operator error?

All the checklists I have seen are stupid long with superfluous lawyer written stuff in like there like "fill O2" and "check for damage"

This creates lengthy checklists which then get ignored or not used. To avoid this I made my own. No its not manufacturer approved but making the checklist actually useable and legible is key to actually using it every time, which is how you stay alive. Manufacturers need to learn 1) how to design out potential assembly/diver errors - e.g all Orings should be contrasting colors so you can immediately see if they are missing or cut. 2) how to write briefly and concisely.

Just last year weren't you lamenting how your scooter couldn't be trimmed in freshwater? Having an un-trimmable scooter released for consumers is an indication of the thoughtfulness of those engineers and honestly I think its an enormous oversight. I wouldn't be surprised if some CCRs have the same types of issues, for instance I have heard that one CCR has 1 absolutely critical oring and a dog hair in it can lead to CO2 bypass. That's the kind of single point failure that needs to be designed out or at least made very difficult to create.

 

[Scubagolf]

Like I said, the ingenuity of fools knows no bounds. Incompatible inhale and exhale connections is a fine idea, so long as WOB and other function and reliability considerations are not harmed. Personally, I find color coding sufficient. When you start worrying about reversing whole CLs and/or BOV or DSV assemblies...you're taking it a bit far.

Product liability will be an issue for devices that require some technical ability but are marketed to any fool. TS&M's vet example is interesting, and makes the point. in life critical situations the manufacturer must design - as far as is possible - to be idiot proof. In most states, product liability based on design or inadequate warnings or instructions is not an issue of negligence, but one of strict liability. In California the plaintiff must prove either that the product failed to meet consumer "expectations", OR that the risks inherent in the design outweigh the benefits of the product as whole. The second test has put a lot of product manufacturers out of business. Again, if you sell a potentially life threatening device to a market with fools you had better make sure the benefits outweigh the risk in using the product.

 

[Dr. Lecter]

Product liability will be an issue for devices that require some technical ability but are marketed to any fool. TS&M's vet example is interesting, and makes the point. in life critical situations the manufacturer must design - as far as is possible - to be idiot proof. In most states, product liability based on design or inadequate warnings or instructions is not an issue of negligence, but one of strict liability. In California the plaintiff must prove either that the product failed to meet consumer "expectations", OR that the risks inherent in the design outweigh the benefits of the product as whole. The second test has put a lot of product manufacturers out of business. Again, if you sell a potentially life threatening device to a market with fools you had better make sure the benefits outweigh the risk in using the product.

Show me one CCR manufacturer who will deliver a unit to anyone who hasn't been certified on it. "As far as is possible" is meaningless word-salad that doesn't demarcate anything. And while this isn't the forum for it, your critique's phrasing suggests to me that you really don't want to argue the finer points of negligence and products liability with me.

 

[gianaameri]

Like I said, the ingenuity of fools knows no bounds. Incompatible inhale and exhale connections is a fine idea, so long as WOB and other function and reliability considerations are not harmed. Personally, I find color coding sufficient. When you start worrying about reversing whole CLs and/or BOV or DSV assemblies...you're taking it a bit far.

Using left-hand thread for one side of the rebreather and right-hand thread for the other side and having little mating notches on the one-way valve holders such that they cannot be reversed is cost effective, does not affect WOB..., and beats color coding (with color coding you can still assemble the rebreather incorrectly).

It is about taking the risk to as low as reasonably practicable (given that the consequences of one human error is death) and you can beat 10:1 color coding with my suggestion (not really my invention, just what I have seen in the Meg and in the POS7).

In any event, does the Hollis Prism 2 have color coding to aide in the assembly (does not seem like it does from the pictures)?

 

[dreamdive]

Here is one: One of our divers encountered a Prism diver who thought that by filling up his scrubber half way is helping him to conserve sorb! Wonder who trained him! So if this guy had died, it was not because of manufacturing error, but because of poor training. Just like "baking" sorb or disturbing the sorb bed otherwise.

I don't think that even the Apoc would be able to remedy this.